This application describes a 3-year training program for the development of an academic career in Infectious Diseases. The principal investigator has completed structured clinical fellowship training at the University of Washington and will expand her research expertise through a unique project that integrates the resources and expertise of several scientists in the Seattle area. The program will promote a broad command of mycobacterial pathogenesis and mechanisms of bacterial persistence and antibiotic resistance operant in the host. Dr. Lalita Ramarishnan will mentor the principal investigator's scientific development in collaboration with Drs. E. Peter Greenberg and Pradeep Singh, two biofilm experts. In addition, an advisory committee of reputable scientists will provide scientific and career advice. Two fundamental problems in the global control of tuberculosis ,(TB) are the ability of the causative organism, Mycobacterium tuberculosis (MTB) to persist for long periods in the host and the long treatment time required for cure. Both impediments are thought to arise from the ability of MTB to achieve a nonreplicating state in vivo, which renders it resistant to killing. Work in other chronic bacterial infections suggests that biofilm formation is responsible for both antimicrobial tolerance and resistance to host mediating killing. Biofilms are multicellular bacterial communities encased in a matrix and the resistance to antimicrobial killing exhibited by biofilm bacteria is thought to be due in part to the accumulation of nonreplicating bacteria therein. I hypothesize that the long duration of therapy required for TB is due to the formation of biofilm-like states in the host. Using a Mycobacterium marinum (MM)-zebrafish model of TB that mimics human disease, I propose to identify non-replicating subpopulations of bacteria in the host and the microenvironments in which they reside. Using a Differential Fluorescence Induction approach, I will identify biofilm-specific genes and determine whether they are expressed in the host and, if so, whether their expression correlates with non-dividing bacterial populations. By creating bacterial strains lacking these biofilm-associated genes, I will determine whether biofilm formation also plays a role in mycobacterial persistence. This will be the first detailed functional analysis of the role of biofilm formation in mycobacterial persistence in vivo.